Aleksi Niemel? <aleksi.niemela / cinnober.com> writes:

> Even if things are implemented intelligently and flexibly in the
> first place we'll be facing few problems. First of all, I'm quite
> sure there's no way we can get the security aspects right with first
> try. If we will, we're probably ending up with way too complex
> product. In any event, this leads to multiple versions just like
> Conrad said, and that imposes the second big problem: early
> adoption.

Just a couple of points.

Firstly, I agree that we won't _rely_ on people having continuously
available net access. However, for those that do, some kind of
intelligent, on-demand update system would be a boon. For those that
don't, exactly the same 'requires' statement could be used to give
them instructions on what they need. I'm not sure I see the need for
multiple versions--this scheme might actually make things simpler.

In terms of security, I agree there are issues here, but not
insurmountable ones. We'll need to involve crypto folks in this.

Part of my interest in this is a longer-term idea. I'd like to see Ruby 
as a dynamic, distributed programming language, running in net
appliances as well as desktop machines. I'd like to see Ruby
applications negotiate and broker services among themselves, providing 
the first real implementation of the roaming agent technologies that
have been hyped over the last few years. I honestly think the
potential is there, but to realize it we need to keep the core of
Ruby small and flexible (hence my 'do it in a library' posts).


Regards


Dave